Go to the bottom of this page. See the search engine and sub-section links.
Go to next page Go to previous page Go to top of this section Go to top page Go to table of contents

Previous Section Headers

User's Guide to NCAR CCM3.6 Search Page


3. CCM3.6 Internals


3.6 Disk and Mass Store File Management

File management includes all activities related to the creation, use, and maintenance of sequential and direct-access files and includes acquiring input files, accessing data from these files, creating output files on disk, and archiving them if applicable.

Several input datasets are required to begin a new case or continue a previous one (see "Model Input Datasets" and "Restart Datasets" ). Often these datasets contain a complete MSS path (e.g. /XXX/yyy/zzz/filename). As a first attempt to find the dataset, the model strips away the MSS part of the filepath and checks if filename exists in the current directory. If the file is not found, the path specified by namelist variable DATADIR is then prepended to the filename and is checked as a full pathname. If the file is still not found, the full pathname of the file as specified in the namelist input is searched on local disk. Finally, if this search is not successful, a subroutine call or shell command will be invoked to retrieve the file from the MSS. Users running on a machine not connected to the NCAR MSS can build the appropriate archival command for their particular site by modifying the few routines containing #ifdef SHELL_MSS CPP directives.

Model output files (see "Model Output Datasets" ) are created in the local directory. Depending on the model's horizontal and vertical resolution, the frequency at which output files are created and the size of the local file system, the aggregate amount of output data created could potentially overflow local disk space. Three control mechanisms exist: (1) control the rate at which data are generated; (2) control the size of the output datasets; and (3) archive the data to an off-line device. Control over the rate at which history and restart data are generated is provided via the namelist variables NHTFRQ and NINAVG. Control over the size of the output datasets is achieved by limiting the number of fields and/or compressing the data before it is written to local disk. Namelist variables EXCLUDE and NDENS provide this functionality. Once an output file is written to the local disk, it may be archived to an offline storage device. If running on an NCAR Cray, a Fortran interface to the local mass storage system is invoked. Otherwise, the cpp token SHELL_MSS may be defined, in which case a shell command is built to do the same job. NCAR archival commands are "mswrite" and "msread". Users at other locations will need to modify the code to build the appropriate command for their local environment. If running on an NCAR Cray, output datasets are removed from local disk after they are archived to the MSS, with the with the exception of the final set of history and restart files. 


Sub Sections

    3.6.1 Disk File Management
    3.6.2 Use of the NCAR Mass Storage System
    3.6.3 Issuing Shell Commands from the Model


 Go to the top of this page. See links to previous section headers.
Go to next page Go to previous page Go to top of this section Go to top page Go to table of contents

 Search for keywords in the CCM3.6 Users GuideSearch page

Questions on these pages can be sent to... erik@ucar.edu .


$Name: ccm3_6_6_latest2 $ $Revision: 1.39.2.1 $ $Date: 1999/03/25 21:38:51 $ $Author: erik $